We often hold the belief that hard work conquers all obstacles, but this notion is misleading.

In any professional endeavor, multiple factors influence the results. The belief that “hard work equals success” is appealing because it emphasizes our ability to control our effort. However, many external variables also shape our paths. For instance, no matter how diligently I train, standing at 5'11" makes pursuing a career in the NBA an uphill battle due to physical limitations.

While hard work can offset some disadvantages, it has its limits. The idea that anyone can achieve anything simply by adjusting their efforts is a fallacy. Our circumstances shape who we are. As Spanish philosopher José Ortega y Gasset famously stated:

“I am I and my circumstance.”

In this discussion, I will explore why data science—or indeed any profession—might not be suitable for everyone, highlighting three key factors that determine our readiness to engage in this field.

Intellectual Capacity — Missing Potential

Artificial intelligence specialists often mention the concept of artificial general intelligence (AGI), referring to systems that possess human-like intelligence. This idea is somewhat perplexing since it treats all humans as a homogeneous group. Regardless of individual intelligence, we are all categorized as having human-level intellect.

However, a closer examination reveals that intelligence varies significantly among individuals. For example, Einstein, often regarded as a paragon of human intelligence, was far more gifted than the average person. While we can broadly categorize humans as having similar intelligence, individual differences become apparent when we focus on specific attributes.

I am not solely discussing IQ; I appreciate Howard Gardner’s theory of multiple intelligences, which suggests intelligence encompasses various skills. One individual may excel in linguistics yet struggle with dancing, while another might have exceptional interpersonal skills but lack self-reflection. Mathematical and logical reasoning represent just one facet of intelligence, and those who are proficient in this area tend to be better equipped for technical fields like data science.

Professionals in technical domains typically possess a unique combination of traits: they are logical, analytical, adept problem-solvers, and exhibit strong critical thinking and abstract reasoning skills. To some extent, these qualities are innate, as we are not born as blank slates; rather, we enter the world with certain predispositions. Those with a natural inclination towards these skills may find themselves more suited for technical disciplines.

Conversely, individuals whose biological predispositions do not favor mathematics and logic may encounter challenges in data science, even with significant effort. Although it is not impossible for them to succeed—given the right environmental influences—such individuals often face steep odds.

Educational Background — Unfulfilled Potential

If intelligence imposes biological limits on our capabilities, education imposes cultural constraints. An analogy can be drawn: intelligence represents the size of our "capability bucket," while education indicates how full that bucket is. Interestingly, many of us underestimate the potential size of this bucket, often because we lack adequate educational experiences.

Our innate capabilities exist, but whether they are realized depends largely on the education we receive. For instance, if both of my parents were engineers, I would likely find it easier to grasp mathematical concepts than my peers. Likewise, a passionate professor teaching a data science course can significantly enhance my learning experience compared to a less engaging instructor. Environmental factors often play a critical role in determining the level of expertise we can attain in any discipline.

Various circumstances can influence the quality of education, including socioeconomic status, historical contexts, the strength of familial bonds, and geographical location. For example, even with engineer parents, residing in New York would afford me better opportunities than living in Detroit, Mumbai, or Nairobi, due to varying access to resources and educational infrastructure. Those unable to access quality educational services face an unjust disadvantage.

Ultimately, dormant intelligence holds no value if it remains unexploited. The presence of potential is meaningless without the proper educational framework and support to nurture and develop those inherent skills. My own journey into AI was facilitated by my father's background in physics, my own studies in aerospace engineering, and exposure to high-level AI courses. Without these experiences, I would have been oblivious to my potential and unlikely to pursue a career in AI.

Knowledge Deficiency — Untapped Potential

Data science has gained immense popularity, often perceived as an easily accessible and lucrative career path. The Harvard Business Review dubbed it the "sexiest job of the 21st century." Numerous online courses suggest that just a few months of dedicated study can equip anyone to secure a data science position at a reputable company. My own experience supports this perception; I began studying AI in September 2017 and landed a job at an AI startup by January 2018—just four months of study, which seems insufficient to master a complex field.

This discrepancy can be explained by recognizing that various learning approaches exist. Had I focused on the theoretical foundations of AI from Turing’s era through early symbolic AI, I would still be grappling with expert systems four years later. Instead, I opted for a brief introductory course on machine learning before diving into deep learning and coding, employing a top-down approach. I prioritized immediate practical applications over establishing a solid theoretical foundation.

However, this method carries risks. Individuals lacking a strong background in mathematics or statistics—unlike my case as an aerospace engineering student—might mistakenly believe that mastering Python, TensorFlow, and a few key models is sufficient for a career in AI, which is a significant misjudgment.

Recently, diverse opinions on this topic have surfaced in Towards Data Science. Soner Y?ld?r?m asserted that data scientists should also grasp data engineering tasks, while Chris The Data Guy controversially claimed that an ML engineer need not have a math background. Sarem Seitz provided an opposing viewpoint just days later.

A trade-off always exists: devoting more time to coding and practical projects can lead to a superficial understanding of theory, while a focus on foundational knowledge may hinder practical skills. Regardless of the path chosen, without adequate time dedicated to learning, some potential will remain untapped. A partial education can jeopardize one’s prospects in the competitive data science and AI job market.

Conclusion

Numerous factors may explain why individuals are ill-suited for certain fields. The three elements discussed—intelligence, education, and knowledge—apply universally across disciplines. Data science, despite its allure and broad applicability, is not immune to these realities.

Some individuals may lack the inherent potential necessary for success in this area, while others may face educational barriers that prevent them from reaching their full capabilities. Additionally, some may overlook essential knowledge required in the field, leading to unutilized potential. Even those with the right intelligence, education, and knowledge might simply prefer not to pursue technical careers. Ultimately, it is clear that certain individuals are better suited for data science than others.

One implication of this situation is that the group of individuals eager to work in data science does not entirely overlap with those who are well-suited for the field. Nonetheless, the intent of this article is not to deter anyone from exploring opportunities in data science or AI but to shed light on the subtle factors that can influence success. A more comprehensive understanding of these dynamics can better prepare us to navigate the challenges on the path to our goals.

This discussion holds relevance as data science is poised to become increasingly prevalent in the future. Its broad applicability may necessitate tech-related skills in many professions that currently seem unaffected. Given the arguments presented here, if data science and AI ultimately displace workers in the near future, the notion that "we can assist individuals in transitioning to new roles created by AI" may not be a viable strategy.

Join me on this journey into the future for more insights on AI, philosophy, and cognitive sciences! Feel free to engage in the comments or connect with me on LinkedIn or Twitter! :)

Further Reading

• Four Mistakes to Avoid While Learning AI
  • AI is wonderful, but we should approach it the right way.
  • towardsdatascience.com
• 5 Reasons Why I Didn’t Take a Master’s Degree in AI/ML/DL
  • And why I studied cognitive science instead.
  • towardsdatascience.com